Summary:
This allows Eradicate to lookup the annotations from the classpath and without requiring the code in the classpath to have been previously analyzed. The benefit is that source files can be analyzed independently of each other as long as the classpath is known.
The main goal is to run be able to run Eradicate as a linter without losing warnings.
We may have to add some more models of the standard libraries as no `Nullable` on a parameter does not necessarily mean that the method does not accept `null`.
Reviewed By: sblackshear
Differential Revision: D6921720
fbshipit-source-id: f525269
Summary: Some tags like `Bucket` are used, but a lot are just added to the list of tags and never read.
Reviewed By: mbouaziz
Differential Revision: D6886980
fbshipit-source-id: 4474d7f
Summary:
There's a lot of code for building up and moving around `Tags`.
When working on cleaning up some of the `Errlog` code, I noticed that `Tags` were included in the JSON and wondered why.
The answer is suprisingly just one thing: only the line tags get used, and even then they are only used to decide what frame to select as the start frame for the trace (i.e., the one that is highlighted first).
That seems like overkill; starting on trace on the actual line where the error occurs, starting at the beginning of the procedure where the error occurs, or starting at the first line of the trace all seem equally reasonable.
If we are happy with any of these alternatives, we can kill `Tags` altogether and potentially save a decent amount space in our JSON artifacts.
Reviewed By: mbouaziz
Differential Revision: D6876752
fbshipit-source-id: 1580127
Summary:
Useful to debug without having to rerun commands.
Looks like this:
```
Usage Error: Failed to execute compilation command:
'/home/jul/infer.fb/infer/bin/../../facebook-clang-plugins/clang/install/bin/clang'
'args.txt'
++Contents of 'args.txt':
'arg1.txt' 'arghorashy_hello_c.txt' 'args.txt'
++Contents of 'arg1.txt':
'arghorashy_c.txt'
++Contents of 'arg_c.txt':
'-c'
++Contents of 'arg_hello_c.txt':
'examples/hello.c'
Error message:
clang-5.0: error: no such file or directory: 'examples/hello.c'
*** Infer needs a working compilation command to run.
```
Reviewed By: mbouaziz
Differential Revision: D6872550
fbshipit-source-id: 65cd026
Summary:
- Combine two fields from ProcAttributes.t into a single field `method_kind` with more information
- New field details whether the procedure is an `OBJC_INSTANCE`, `CPP_INSTANCE`, `OBJ_CLASS`, `CPP_CLASS`, `BLOCK`, or `C_FUNCTION`
- `is_objc_instance_method` and `is_cpp_instance_method` fields no longer necessary
- Changed `is_instance` field in CMethod_signature to `method_kind` field of type ProcAttributes.method_kind
Reviewed By: dulmarod
Differential Revision: D6884402
fbshipit-source-id: 4b916c3
Summary: If the procedure is defined, then the attributes should be the same on the specs files or on the attributes table.
Reviewed By: sblackshear
Differential Revision: D6910086
fbshipit-source-id: 709b290
Summary:
- small optimization by starting deconstructing procnames/types in the dispatcher rather than the matchers
- as a consequence, returns fast for unhandled constructs like Java procnames or types
- Java is still not handled but at least does not crash
- re-enable Inferbo for Java
Reviewed By: jberdine
Differential Revision: D6912304
fbshipit-source-id: 76e95a8
Summary:
In some cases infer runs clang commands where we do not attach the plugin,
notably for pre-processor-only commands like `clang -E foo.c`. Usually these
commands are used by build systems to test their output, so infer should not
swallow it.
This makes infer echo the stdout of clang whenever we do not attach the plugin
to clang. When the plugin is attached, stdout is captured because that's where
the plugin outputs its results, so we cannot do the same in this case (not that
we would want to yet).
Fixes#696.
Reviewed By: mbouaziz
Differential Revision: D6912101
fbshipit-source-id: c4ad2e4
Summary:
Record "capture phases" in the runstate and in the source files table of the
database. Use this instead of filesystem timestamps to decide which files need
re-analyzing in the reactive analysis.
Reviewed By: jeremydubreil
Differential Revision: D6760833
fbshipit-source-id: 7955621
Summary:
The boolean lock domain is simple and surprisingly effective.
But it's starting to cause false positives in the case where locks are nested.
Releasing the inner lock also releases the outer lock.
This diff introduces a new locks domain: a map of locks (access paths) to a bounded count representing an underapproximation of the number of times the lock has been acquired.
For now, we just use a single dummy access path to represent all locks (and thus a count actually would have been sufficiently expressive; we don't need the map yet).
But I'm planning to remove this limitation in a follow-up by refactoring the lock models to give us an access path.
Knowing the names of locks could be useful for error messages and suggesting fixes.
Reviewed By: jberdine
Differential Revision: D6182006
fbshipit-source-id: 6624971
Summary: There's a new `ocaml_pos` type that the other clang frontend exceptions use, but Self.SelfClassException still used the raw tuple. Now, SelfClassException also uses this type.
Reviewed By: dulmarod
Differential Revision: D6900258
fbshipit-source-id: 94c7042
Summary:
- During backend execution, infer will log detailed stats about procedure analysis
- Logging is integrated with EventLogger
- `events_to_log` field added to Stats.t record in InferPrint
- New format in InferPrint - Logs
- `format_list` type changed to have a Utils.Outfile option to support Logs format
Reviewed By: dulmarod
Differential Revision: D6834538
fbshipit-source-id: 8c847f5
Summary:
Previously, we could understand than an access was safe either because it was possibly owned or protected by a thread/lock, but not both. If an access was both protected by a lock and rooted in a paramer (i.e., possibly owned), we would forget the ownership part of the precondition and remember only the lock bit. This leads to false positives in cases where an access protected by a lock is owned, but another unowned access to the same memory is not protected by a lock (see the new `unownedLockedWriteOk` E2E test for an example).
This diff makes access safety conditions disjunctive so we can simultaneously track whether an access is owned and whether an access is protected by a thread/lock. This will fix false positives like the one explained in T24015160.
Reviewed By: jberdine
Differential Revision: D6671489
fbshipit-source-id: d17715f
Summary:
- ast_node argument is now optional in functions `unimplemented` and `incorrect_assumption` in cFrontend_config
- The argument type was already an option, and the majority of the calls were with 'None'. This makes the function more intuitive to use
Reviewed By: sblackshear
Differential Revision: D6846141
fbshipit-source-id: 13deb8f
Summary:
We already knew not to warn when non-resource `Closeable`'s like `ByteArrayOutputStream` weren't closed, but we still warned on their subtypes.
This diff fixes that problem by checking subclasses in the frontend.
This also removes the need for Java source code models of non-resource types, so I removed them.
Reviewed By: jeremydubreil
Differential Revision: D6843413
fbshipit-source-id: 60fe7fb
Summary: The heuristics to determine the end of a block/procedure was too brittle, the new one ignores non significant instructions.
Reviewed By: jvillard
Differential Revision: D6845380
fbshipit-source-id: feab557
Summary:
The infer results directories in buck-out/ are "cleaned up" to avoid polluting
the Buck cache with too much data or non-deterministic data. In particular, the
runstate is deleted, which confused subsequent infer processes trying to read
the pre-existing results directory.
Add a special case in infer to delete pre-existing results directories in
buck-out instead of trying to load their state.
Reviewed By: mbouaziz
Differential Revision: D6845128
fbshipit-source-id: 5c716aa
Summary:
- `NonZeroInt` for added guarantees on the invariants of `SymLinear` coefficients
- some simplifications
- some optimizations
Reviewed By: jvillard
Differential Revision: D6833968
fbshipit-source-id: 39e28a0
Summary:
Make dead code detection part of `make test` so that dead code stops creeping
in. It's only enabled if all the analysers are enabled and if this is a
facebook build, because the dead code detection will have false positives
otherwise.
Reviewed By: mbouaziz
Differential Revision: D6807395
fbshipit-source-id: ebbd835
Summary:
This diff fixes the translation of `new` and `placement new` with one argument. If `placement new` has more than one argument it means that it is user-defined (this will be addressed in another diff).
update-submodule: facebook-clang-plugins
Reviewed By: sblackshear, mbouaziz
Differential Revision: D6807751
fbshipit-source-id: 7cf0290
Summary:
Before this diff, the `Ondemand` module would not cache the results of the function `analyze_proc_desc`, which is used by the toplvel iteration.
This should not have any effect on the performances at this point as the summaries were already cached in the `Specs` module. Now, we can start remove the use of the cache in the `Specs` module to avoid the duplication. Caching at the level of `Ondemand` is better as we can safely cache the information that the outcome of the analysis is `None`, which avoids scanning the filesystem or the DB multiple times.
Reviewed By: mbouaziz, jvillard
Differential Revision: D6713546
fbshipit-source-id: 309701b
Summary: This should allow to report several occurences of the an issue appearing several times within the same method.
Reviewed By: jvillard
Differential Revision: D6783298
fbshipit-source-id: 5555906
Summary: It was getting a bit difficult to tell which functions belonged where.
Reviewed By: jvillard
Differential Revision: D6764764
fbshipit-source-id: f9faada
Summary:
This lets us fix the limitation of reporting false positives when a `private` function calls `build()` on a parameter without passing all of the required props.
We will now report such issues in the caller only if it fails to pass the required props.
An unfortunate consequence of this change is that we lose track of where the actual call to `build` occurs--we now report on the declaration of the caller function rather than on the call site of `build`.
I'll work on addressing that in a follow-up.
Reviewed By: jeremydubreil
Differential Revision: D6764153
fbshipit-source-id: 3b173e5
Summary:
The captured variables of a closure are tuples (id, var, typ) with the implicit assumption
that &var -> id holds in the heap. This is true when the closure is created, but is not enforce otherwise.
This becomes a problem when the closure is stored in the heap, goes passed a bi-abduction, and then it's executed
(see new test). This was failing before this diff and now succeeds.
We add the verification of this constraint to the normalization of sigma.
At the moment I expect Precondition_not_met to be removed, but also later, we will be able to compute retain cycles
over the closures, as the correct captured variable info is kept through the execution.
Reviewed By: jvillard
Differential Revision: D6796525
fbshipit-source-id: a8a7655
Summary:
Not sure what an "iCFG" is but the dotty is only about CFGs anyway.
Diff obtained by mass-`sed`.
Reviewed By: sblackshear
Differential Revision: D6324280
fbshipit-source-id: b7603bb
Summary:
Also make it optional, since it's only used for debug messages. Name a couple
more of these for other similar functions.
Reviewed By: sblackshear
Differential Revision: D6797385
fbshipit-source-id: e6e9b2e
Summary:
I needed to do this for something, now I don't know if I want to do the thing
anymore but this seems generally useful to decrease a little bit the size of
Config.ml.
Reviewed By: sblackshear, mbouaziz
Differential Revision: D6796427
fbshipit-source-id: d9c009d
Summary:
Also, always log failures.
This also shows that the dead code detection does not detect dead exceptions :/
Reviewed By: jeremydubreil
Differential Revision: D6796843
fbshipit-source-id: 3d0ff5c
Summary:
Also, make it explicit when we load the global tenv instead of the per-file tenv.
This allows for some nice simplifications in some places, notably:
- `tenv_file` is gone from `Exe_env.file_data`
- `DB.global_tenv_fname` is no more
This will help moving the tenv from the capture/source_file/ directories on the
filesystem to the database, as keys for the relevant table are `SourceFile.t`.
Reviewed By: mbouaziz
Differential Revision: D6796594
fbshipit-source-id: 1ffd5b0
Summary:
The custom exception does not appear to serve any purpose. Also refactor the
complaining code to avoid duplication.
Reviewed By: mbouaziz
Differential Revision: D6784945
fbshipit-source-id: a2d969b
Summary:
They were constructed for each source file, and then joined into a global call
graph, only to get per-file lists of procedures. A tad wasteful.
Get this list from cfgs instead. Still record them in `exe_env` for now as
changing that code is a whole other beast.
One test falls victim of the flakiness of the analysis of recursive functions.
Reviewed By: jeremydubreil, mbouaziz
Differential Revision: D6324268
fbshipit-source-id: d5ff58f
Summary:
Last piece of significant logic relying on the call graph: the execution
environment. Mostly carve out the logic to detect duplicate symbols to not rely
on the call graph.
Also make the keys of the `file_map` be source files, because not having the
"cg filename" makes it harder otherwise.
Reviewed By: sblackshear
Differential Revision: D6621196
fbshipit-source-id: cab50ba
Summary:
In preparation for getting rid of call graphs, we need to find another way to
get the list of defined procedures (which is the only place where we use the
globally-computed call graph for now).
The natural way to get the list of procedures defined in a file is to load the
cfg for that file and look at the proc names that are the keys of the cfg. This
is way too expensive, as the CFG is big. Thus, we cache this list of proc names
as another column in the SQLite database of cfgs. This gives good performance
in benchmarks.
Reviewed By: jeremydubreil
Differential Revision: D6621142
fbshipit-source-id: ed265fe
Summary: At each call to `Component$Builder.build()`, checks that the required props for `Component` have been set via prior calls to the `Builder`. Does not yet handle `Prop(optional = true)`, but will address that in a follow-up.
Reviewed By: jeremydubreil
Differential Revision: D6735524
fbshipit-source-id: 0c812fd
Summary:
Record the db schema, infer version, and run dates into
infer-out/.infer_runstate.json. This allows us to check on startup whether the
results directory was generated using a compatible version of infer or not, and
give a better error message in the latter case than some SQLite error about
mismatching tables.
This will be used in a follow-up diff to record capture phases too, and avoid
relying on filesystem timestamps of the infer-out/capture/foo/ directories for
reactive analysis.
Had to change some tests Makefiles to make sure they do not attempt to re-use
stale infer-out directories, which would now fail the run.
The stale infer-out directory gets deleted if `--force-delete-results-dir` is
passed (but a warning still gets printed).
Reviewed By: mbouaziz
Differential Revision: D6760787
fbshipit-source-id: f36f7df
Summary:
This declutters `CommandLineOption` a little bit, and will be useful in a
follow-up diff where `InferCommand.t` will be used from an atd-generated file.
Reviewed By: mbouaziz
Differential Revision: D6772990
fbshipit-source-id: 3d32d00
Summary:
Previously imports with relative filenames would not get resolved so the result
would depend on where infer had been run from. Usually this was the project
root. Now, resolve path names of imports relative to the file doing the
`#IMPORT`. This changes behaviour most of the time.
Reviewed By: dulmarod
Differential Revision: D6784740
fbshipit-source-id: 4ccb7bf
Summary:
This changes the syntax for AL imports from `#IMPORT <file>` to `#IMPORT
"file"`. As a side-effect, the `file` part is now lex'd more permissively too.
Reviewed By: dulmarod
Differential Revision: D6784669
fbshipit-source-id: cc1bb73
Summary:
- During program translation, infer logs details about SelfClassException exceptions that are caught
- Logging is integrated with EventLogger library, uses existing FrontendException event type
- ast_node field in FrontEndException record used to store SelfClassException class_name field
- SelfClassException exception type extended to add support for storing exception details
- All instances where SelfClassException exception is raised modified to pass these details
Reviewed By: mbouaziz
Differential Revision: D6760513
fbshipit-source-id: a8efa9d
Summary:
- During program translation, infer logs details about IncorrectAssumption exceptions that are caught
- Logging is integrated with EventLogger library, uses existing FrontendException event type
- IncorrectAssumption exception type extended to add support for storing exception details
- All instances where IncorrectAssumption exception is raised modified to pass these details
Reviewed By: mbouaziz
Differential Revision: D6759287
fbshipit-source-id: 64f520e
Summary:
Reading a different .inferconfig makes little sense and is in contradiction
with the fact that we try hard to make all the infer processes agree on the set
of options.
Reviewed By: sblackshear
Differential Revision: D6761146
fbshipit-source-id: 67a0c54
Summary:
- During program translation, infer logs details about Unimplemented exceptions that are caught
- Exception type, source location, exception triggering location, and the ast_item it occurred in
- Logging is integrated with EventLogger library
- New type of event in EventLogger
- Unimplemented exception type extended to add support for storing these extra details
- All instances where unimplemented exception is raised modified to pass these details
Reviewed By: dulmarod
Differential Revision: D6748734
fbshipit-source-id: 725c7f3
Summary:
and add mli. We already had the logic for iterating over call chains, but it was overfitted to the should-update analysis.
Will use the generalized version in a follow-up.
Reviewed By: jvillard
Differential Revision: D6740692
fbshipit-source-id: 8c0d89f
Summary:
Getting double-logged messages when logs have not yet been setup is confusing,
so just don't log these messages to a log file.
Reviewed By: mbouaziz
Differential Revision: D6739173
fbshipit-source-id: 14db6b0
Summary:
Was trying to decide where to add a new Java utility function and realized that things are a bit disorganized.
Some operations on `Typ.Name.t`'s live in `Typ.Procname`, and some live inside an inner `Java` module whereas some are outside of the module with a `java_` prefix.
Let's move toward putting all Java/C/Objc/C++-specific functions in dedicated modules.
This diff does some of the work for Java.
There are Java-specific functions that operate on `Typ.Procname.t`'s that will have to be converted to work on `Typ.Procname.Java.t`'s, but changing those clients will be more involved.
Will also move C/Objc/C++ functions in a follow-up.
Reviewed By: jeremydubreil
Differential Revision: D6737724
fbshipit-source-id: cdd6e68
Summary: Use the Hashtbl functions directly as `Cfg` knows that a cfg is a hashtbl.
Reviewed By: sblackshear, jeremydubreil
Differential Revision: D6727732
fbshipit-source-id: 2cdda91
Summary:
`&::.*-->` allows to match any path end.
Used for models of `std::array` to force unmodelled functions (and types) to have a Skip summary
Depends on D6408415
Reviewed By: jvillard
Differential Revision: D6611203
fbshipit-source-id: 6663b2c
Summary:
In buck queries, the `regex` in `kind(regex, ...)` is open, it can match superstrings, like `prebuilt_cxx_library` when we want `cxx_library` instead.
This adds `^` and `$` to our existing kind filtering.
Also optimizes the query in `buck_target_determinator.py`.
Reviewed By: dulmarod
Differential Revision: D6722665
fbshipit-source-id: 22d839f
Summary:
Horrible but somewhat-documented hack to get inter-file dead code analysis for
the OCaml code.
Reviewed By: mbouaziz
Differential Revision: D6724234
fbshipit-source-id: 3a5b9cd
Summary:
Found the dead code with the script in the next commit, iteratively until no
warnings remained.
Methodology:
1. I kept pretty-printers for values, which can be useful to use from infer's REPL (or
when printf-debugging infer in general)
2. I kept functions that formed some consistent API (but not often, so YMMV), for instance if it looked like `Set.S`, or if it provides utility functions for stuff in development (mostly the procname dispatcher functions)
3. I tried not to lose comments associated with values no longer exported: if the value is commented in the .mli and not the .ml, I moved the comment
4. Some comments needed updating (not claiming I caught all of those)
5. Sometimes I rewrote the comments a bit when I noticed mis-attached comments
Reviewed By: mbouaziz
Differential Revision: D6723482
fbshipit-source-id: eabaafd
Summary:
Almost all the files are supposed to do this. When they don't, the build adds
it automatically anyway, but it's better to always include it to avoid
confusion (and help other automated tools).
Reviewed By: mbouaziz
Differential Revision: D6723285
fbshipit-source-id: 0fe8a16
Summary:
- After completing program translation, infer logs # of attempted procedure translations and # of completed procedure translations via EventLogger
- New possible type of event in EventLogger
Reviewed By: dulmarod
Differential Revision: D6711662
fbshipit-source-id: 5e31332
Summary:
- After completing program translation, infer logs # of attempted procedure translations and # of completed procedure translations
- New global state in cFrontend_config.ml to represent # of attempted procedure translations and # of failed procedure translations
- Will be integrated with logging framework
Reviewed By: dulmarod
Differential Revision: D6703248
fbshipit-source-id: 10c916a
Summary:
In Java, static variables are distinguished by package/class:
the file where they are defined doesn't matter.
Fixes#831.
Closes https://github.com/facebook/infer/pull/833
Reviewed By: jeremydubreil
Differential Revision: D6661240
Pulled By: sblackshear
fbshipit-source-id: beeb2f9
Summary: This should make no difference as the `Ondemand` would already only run the analysis when the procedure description is found, and naturally skip the analysis otherwise.
Reviewed By: sblackshear, jvillard
Differential Revision: D6705813
fbshipit-source-id: 44bffee
Summary:
This makes it easier to add more .atd files in the future: just add the
filename to `INFER_ATDGEN_STUB_BASES`.
Reviewed By: mbouaziz
Differential Revision: D6711695
fbshipit-source-id: 0d88daa
Summary:
Some commands (mostly `infer report`) would attempt to run the initialisation
code of infer from the default results directory instead of the one used by the
test. This is mostly harmless because we do not actually use anything from the
directory (typically, we pass `--from-json-results foo.json` and only foo.json
matters). However, this can trip the initialisation code, eg on db schema
changes.
Reviewed By: mbouaziz
Differential Revision: D6711641
fbshipit-source-id: f04b4c7
Summary: The models tenv is loaded inside a global variable that is never used.
Reviewed By: jeremydubreil
Differential Revision: D6692656
fbshipit-source-id: 4166a22
Summary:
This commit augments codes to clean up temporary files generated by the clang frontend.
Currently clang frontend leaves large number of temporary files int the tmp directory, and it could be seen for example by running these command:
```
git clean -xdf infer/models/
mkdir /tmp/infer
env TMPDIR=/tmp/infer make infer_models
ls -l /tmp/infer
```
P.S. Analyzing real project could easily cause each infer capture to leave hundreds of files in /tmp
P.P.S. There are 11 total references to Filename.temp_file however, other 9 seems don't leak temporary files in such large scale (at least not when using the clang frontend).
Closes https://github.com/facebook/infer/pull/816
Reviewed By: jvillard
Differential Revision: D6385311
Pulled By: dulmarod
fbshipit-source-id: f7956b0
Summary:
This is a more accurate type since an execution environment is always a single
file, and allows us to make one of the fields of `Exe_env.t` immutable.
Reviewed By: mbouaziz
Differential Revision: D6620477
fbshipit-source-id: 9553516
Summary:
This avoids relying on the directories in infer-out/captured/ being created,
and instead gets the list of captured source files from the DB. This gives a
better type to clusters: `SourceFile.t` instead of `DB.source_dir`, which makes
the code a bit nicer too.
Reviewed By: jeremydubreil
Differential Revision: D6620460
fbshipit-source-id: c0edbf6
Summary: Get the error message from the database when there's an error, together with the error type.
Reviewed By: mbouaziz
Differential Revision: D6621695
fbshipit-source-id: 6bc706d
Summary: Somehow sqlite allows this simpler statement when using `exec` but not `prepare`.
Reviewed By: mbouaziz
Differential Revision: D6407732
fbshipit-source-id: 01f029f
Summary:
This makes sure that sqlite doesn't hold read locks for longer than necessary,
which could starve the process of cleaning up the WAL file. This ensures that
the statement is reset as soon as we're done reading.
I haven't observed a difference with this change, and could not find evidence
that it should change something in the docs. Internet wisdom pointed at this as
a potential issue and I was observing it in another change, so it's good to
rule it out.
Reviewed By: mbouaziz
Differential Revision: D6404353
fbshipit-source-id: a123cd6
Summary: Previously we had a single sanitizer kind for escaping, but this isn't quite right. A function that escapes a URL doesn't necessarily make a string safe to execute in SQL, for example.
Reviewed By: the-st0rm
Differential Revision: D6656376
fbshipit-source-id: 572944e
Summary: This should avoid making copies of procedure descriptions which are mutable data-stuctures.
Reviewed By: sblackshear
Differential Revision: D6658527
fbshipit-source-id: 688a142
Summary: The checker should only propagate the nullablility on the lhs when of pointer type.
Reviewed By: sblackshear
Differential Revision: D6630294
fbshipit-source-id: 07fe3d6
Summary: There was several implementations of the same function accross the codebase
Reviewed By: sblackshear
Differential Revision: D6658266
fbshipit-source-id: e12507b
Summary: This subdirectory was only containing tests related to nullable on Objective C.
Reviewed By: sblackshear
Differential Revision: D6657654
fbshipit-source-id: 11003f2
Summary: Filtering the defined procedure at this level is not necessary. This check already happens when running the analysis in Ondemand. This could also cause flakiness if the "definedness" here does not agree with the check done in Ondemand. The fact that the analysis of a procedure is triggered from the top-level iteration or on-demand when analyzing another procedure is not deterministic.
Reviewed By: sblackshear
Differential Revision: D6575057
fbshipit-source-id: ff0bc2d
Summary:
Upgrade ocamlformat to 0.3, and (necessarily) base to v0.10.0.
- Fix accumulated mis-formatting
- Update opam.lock to unbreak clean build
- Update to base v0.10.0
- Update opam.lock for base
- Update offline opam repo
- Everyone should already have removed their ocamlformat pin
- ocamlformat 0.3 supports output to stdout natively
- bump version of ocamlformat
Reviewed By: jeremydubreil
Differential Revision: D6636741
fbshipit-source-id: 41a56a8
Summary:
Model for `folly::split` that handles the representation in the cpp model.
Depends on D6544992
Reviewed By: jvillard
Differential Revision: D6545006
fbshipit-source-id: 2b7a139
Summary:
Before this diff, the nullable checker would not be able to find annotations involving methods annotated in the protocols
update-submodule: facebook-clang-plugins
Reviewed By: sblackshear
Differential Revision: D6534893
fbshipit-source-id: 39bd3dd
Summary:
Allows:
- matching function arguments with or without capturing,
- capturing part of an argument, e.g. expression only,
- optional arguments, wrapped into an OCaml option if captured.
Reviewed By: jvillard
Differential Revision: D6544992
fbshipit-source-id: a64ba45
Summary: This is to allow the bi-abduction analysis and the nullable checker for Clang languages to run together without stepping on each other toes.
Reviewed By: sblackshear
Differential Revision: D6567934
fbshipit-source-id: a318c33
Summary: This factors out some duplicated code for {,de}serializing source files.
Reviewed By: mbouaziz
Differential Revision: D6324234
fbshipit-source-id: 1741657
Summary:
Instead of storing the cfgs of source files inside their own individual files,
put them in results.db, in their own table. (that table may change in the
future to map source files to more than just their cfgs, eg their tenv as well)
Reviewed By: jberdine
Differential Revision: D6297201
fbshipit-source-id: 7fa891d
Summary: There was a back and forth conversion between `string` and `IssueType.t` which was not necessary.
Reviewed By: sblackshear
Differential Revision: D6562747
fbshipit-source-id: 70b57a2
Summary:
This diff adds a layer of report deduplication logic in addition to
the existing scheme.
Suppose issue 1 with trace1a and trace1b, and issue 2 with trace2a and
trace2b. If trace1a ends at the same location as trace2a (resp.,
trace2b) and trace1b ends at the same location as trace2b (resp.,
trace2a), then consider issues 1 and 2 to be duplicates.
This chooses to report the issue with the smaller sum of trace
lengths, breaking ties using the issue hashes, and eventually the
entire issue. Therefore there is a potential for flakiness with
respect the the choice of which report to make within a
hash-equivalence class.
Reviewed By: sblackshear
Differential Revision: D6519607
fbshipit-source-id: 63210ab
Summary: As Dulma pointed out, adding or removing paramters in a method in Objective C is changing the name of the method. Such changes should not make pre-exisiting issues reported as introduced. This diff is to prevent this by only keeping in the bug hash the part of the name that is before the first colon.
Reviewed By: dulmarod
Differential Revision: D6491215
fbshipit-source-id: 3c00fae
Summary: It is difficult to understand a lock consistency violation if error message includes an access path with a logical variable or a temporary variable as a base. As a temporary fix, we want to suppress all such warnings.
Reviewed By: sblackshear
Differential Revision: D6460559
fbshipit-source-id: 6f3fc18
Summary:
Our model of unique_ptr and shared_ptr relied on the fact that we could C-style cast a pointer to the internal pointer type used in the smart pointer.
This is wrong when the smart pointer is used with a custom deleter that declares its own pointer type whose is not constructible from just a single pointer.
Reviewed By: dulmarod
Differential Revision: D6496203
fbshipit-source-id: 1305137
Summary: Local `CKComponentScope`'s are often created purely for their side effects, so it's fine for them to be unread.
Reviewed By: jeremydubreil
Differential Revision: D6475475
fbshipit-source-id: 17e869a
Summary: This would allow the checker to detect indirect nullable violations, i.e. violations that are involving intermediate method calls on potentially `nil` values.
Reviewed By: sblackshear
Differential Revision: D6464900
fbshipit-source-id: 3663729
Summary: NSDictionary initialization will crash when using `nil` as a key or as a value
Reviewed By: dulmarod
Differential Revision: D6466349
fbshipit-source-id: 57bb012
Summary: For the Buck integration for Java, caching when a summary is not found avoids going through the whole classpath every time a summary for the same method is not found.
Reviewed By: mbouaziz, jvillard
Differential Revision: D6402833
fbshipit-source-id: 4feb422
Summary: This will avoid collisions when the inner classes are implementing the same methods. For example, the previous version of the bug hash could conflate the issues when several annonymous inner classes are implementing the same method, e.g. a annonymous subclass of `Runnable` implementing `run()`.
Reviewed By: sblackshear
Differential Revision: D6461594
fbshipit-source-id: 2bb8545
Summary: I always get confused by `accessPath.ml` not being next to HIL when trying to open files
Reviewed By: sblackshear
Differential Revision: D6462980
fbshipit-source-id: 8ba9b71
Summary:
Simpler bug hash that is more independent of the underlying analysis. This now computes the hash based on:
- the bug type.
- the base filename: i.e for my/source/File.java, just keep File.java. So the hash will not change when moving files around.
- the simple method name: i.e. without package information and list of parameters. So changing the list of parameters will not affect the bug hash.
- the error message were the line numbers have been removed. So moving code or reformatting will not affect the hash.
Reviewed By: jberdine
Differential Revision: D6445639
fbshipit-source-id: 82e3cbe
Summary:
Summaries can be big, and they can always be printed via `infer report` if we want to see them.
There's no reason to log them eagerly, even in debug mode.
Reviewed By: jeremydubreil
Differential Revision: D6451815
fbshipit-source-id: 643cd47
Summary: In every place this was used except one, `debug_mode` is also used as a gate.
Reviewed By: jeremydubreil
Differential Revision: D6450913
fbshipit-source-id: 6a5716d
Summary: To avoid false positives, we treat `operator[]` in cpp as container read. Moreover, if a container `c` is owned, we make all accesses `c[i]` to be also owned.
Reviewed By: sblackshear
Differential Revision: D6396574
fbshipit-source-id: 94aabff
Summary:
On spinning disks the performance of commits are worse when `synchronous=NORMAL`.
Reading the documentation of SQLite, when `synchronous=OFF` there's a risk of DB corruption when the operating system crashes or the computer loses power before that data has been written to the disk surface; on the other hand, a crash in Infer should keep data in the DB in a sound state.
Buck reached the same conclusions too: 4680162279
Reviewed By: mbouaziz
Differential Revision: D6413384
fbshipit-source-id: 99e4650
Summary:
It seems that the abstraction instructions were not previously added the the CFG.
This is a functional changes to make sure that the abstraction state is always added. We can simplify the code later and just run this step before storing the CFG instead of after loading them.
Reviewed By: sblackshear, jvillard
Differential Revision: D6383672
fbshipit-source-id: cedcb8a
Summary:
Deduping issues when generating a single report and then diffing the
reports can lead to introduced issues being considered duplicates of
existing issues.
Reviewed By: sblackshear
Differential Revision: D6414673
fbshipit-source-id: bba81fd
Summary:
As da319 points out, we did not handle this case correctly before. There were a few reasons why:
(1) An assignment like `struct S s = mk_s()` gets translated as `tmp = mk_s(); S(&s, tmp)`, so we didn't see the write to `s`.
(2) We counted uses of variables in destructors and dummy `_ = *s` assignments as reads, which meant that any struct values were considered as live.
This diff fixes these limitations so we can report on dead stores of struct values.
Reviewed By: da319
Differential Revision: D6327564
fbshipit-source-id: 2ead4be
Summary:
justmovingthingsaround
Models need these functions, they have to be somewhere else.
The split might seem weird for now but will (hopefully) look more obvious in the following diff.
Reviewed By: skcho
Differential Revision: D6408322
fbshipit-source-id: c7e430f
Summary:
Extends `ProcnameDispatcher` to allow matching typenames only.
There isn't much new here, mainly moving stuff so that we only have to open one module to use the operators.
Reviewed By: skcho
Differential Revision: D6408245
fbshipit-source-id: afc6533
Summary: I accidentally deleted the support for `infer report file.specs` which was printing the summary to standard output.
Reviewed By: sblackshear
Differential Revision: D6416690
fbshipit-source-id: 62246f3
Summary:
The diff is very big but it's mostly removing code. It was inspired by the fact that we were getting Dead Store FPs because we were modeling some functions from CoreFoundation and CoreGraphics directly as alloc in the frontend, which caused the parameters of the function to be seen as dead. See the new test.
To deal with this, if we are going to skip the function, we model it as malloc instead. Given how many models we had for those "model as malloc" functions, I removed them to rely solely on the new mechanism.
The modeling of malloc and release was still based on the old retain count implementation, even though all we do here is a malloc/free kind of analysis. I also changed
that to be actually malloc/free which removed many Assert false in the tests. CFRelease is not exactly free though, and it's possible to use the variable afterwards. So used a custom free builtin that only cares about removing the Memory attribute and focuses on minimizing Memory Leaks FPs.
Otherwise we were translating CFBridgingRelease as a special cast, and this wasn't working. To simplify this as well, I removed all the code for the special cast, and just modeled CFBridgingRelease and CFAutorelease also as free_cf, to avoid Memory Leak false positives. I also treated the cast __bridge_transfer as a free_cf model. This means we stopped trying to report Memory Leaks on those objects.
The modeling of CoreGraph release functions was done in the frontend, but seemed simpler to also simplify that code and model all the relevant functions.
Reviewed By: sblackshear
Differential Revision: D6397150
fbshipit-source-id: b1dc636
Summary:
This is a good moment to close Sqlite's DB handles, and in general can be used to postpone some actions right before infer terminates.
Since exiting is done via uncaught exception handling, the `late_epilogue` callback will run at the very end, even after all the `at_exit` callbacks have been invoked. The only exception is made in case of signalling, in which case the `late_epilogue` is still invoked, but before any of the `at_exit` callbacks.
Reviewed By: jvillard
Differential Revision: D6404961
fbshipit-source-id: 8ff7a05
Summary:
The model is the same as `com.google.common.base.Preconditions`.
We could imagine a more generic ways of dealing with `x.y.Z.checkNotNull()` but this would work for now.
Reviewed By: sblackshear
Differential Revision: D6341869
fbshipit-source-id: 5b6e507
Summary:
- Plug model checkers
- Add alloc size safety condition on alloc of negative, zero or big size
Reviewed By: sblackshear
Differential Revision: D6375144
fbshipit-source-id: bbea6f3
Summary:
A modeled function is not only an evaluator but also a checker, at least in Inferbo where both things happen in two passes.
This diff just prepares for it without generating new alarms.
Reviewed By: jvillard
Differential Revision: D6373051
fbshipit-source-id: 264696f
Summary:
In C++ some modeled functions have definitions, which leads to traces
that contain an access from the modeling, but continue on into the
implementation of the modeled function. Such traces appear the same as
those that are truncated due to limitations of the buck integration in
the Java analysis. Since all Java models are for functions without
definitions in the code base, this diff limits the truncated trace
suppression to the Java analysis.
Reviewed By: sblackshear
Differential Revision: D6373793
fbshipit-source-id: 1f01509
Summary: There is a lot of code to create LaTeX output of the Infer datastructures, but this does not seem to be used anymore.
Reviewed By: jvillard
Differential Revision: D6355686
fbshipit-source-id: 55de8e9
Summary:
This field was always empty.
depends on D6351097
Reviewed By: sblackshear, jvillard
Differential Revision: D6351243
fbshipit-source-id: 4a74bea
Summary: This option was for compatibility with the command line options of the previous, but is no longer used. This diff removes the option and the deprecated code.
Reviewed By: sblackshear, mbouaziz
Differential Revision: D6351097
fbshipit-source-id: 0e4cfc5
Summary: This will avoid confusions when running `-a infer --racerd` which would silently not running RacerD before this diff.
Reviewed By: sblackshear
Differential Revision: D6374139
fbshipit-source-id: 2cb5004
Summary: Adding a null key or a null value will cause a runtime exception.
Reviewed By: sblackshear
Differential Revision: D6378618
fbshipit-source-id: 8bd27c6
Summary:
This resolves#796 . Effectively it adds file specific suffix to name of all global initializers (so initializersof two global variable of the same name will have unique Typ.Procname). which is the same rule as currently used by constructing Procname for the static functions. However this change applies to initializers of all global variables and not just static (arguably it's a right thing. since GCC used to allow multiple global variables with the same name).
Consequences of this change that it becomes impossible to know name of generated initialization function of global ('extern') variables. However get_initializer_pname function is only referenced by the frontend (when creating initializer for the defined global variables) and by the SIOF checker.
Closes https://github.com/facebook/infer/pull/801
Reviewed By: jvillard
Differential Revision: D6335034
Pulled By: dulmarod
fbshipit-source-id: 1a92c08
Summary:
Allow capturing function arguments.
Model functions don't have to match on a list any more.
Depends on D6347829
Reviewed By: jvillard
Differential Revision: D6350628
fbshipit-source-id: e88b758
Summary: When not matching overloads, when the wrong number of arguments is given, instead of just no matching the function, we may want to fail, e.g. for internal-use functions.
Reviewed By: jvillard
Differential Revision: D6347829
fbshipit-source-id: 48f41be
Summary:
This was already dead code that didn't know it was dead, doubly so:
1. Only active with `-a biabduction`, which is deprecated
2. Doesn't do anything since it somehow always iterates over an empty list of procedures (I don't really know why that is, but testing shows this is the case)
Reviewed By: jeremydubreil
Differential Revision: D6348430
fbshipit-source-id: 230d05d
Summary:
Naming a variable `_foo` makes the compiler not warn about them if they are
unused, but there are lots of instances of such variables in the code where
they are in fact used, defeating the warning and introducing confusion for
those used to this naming convention.
Basically `sed -i -e "s/ _\([a-zA-Z][a-zA-Z0-9_']*\)/ \1_/g" **/*.ml` followed
by manual fixing of compilation errors (lots of `compare__foo` ->
`compare_foo_`).
Reviewed By: mbouaziz
Differential Revision: D6358837
fbshipit-source-id: 7ffb4ac
Summary: Adding a nil object to an NSArray will crash. Adding this case to the checker.
Reviewed By: sblackshear
Differential Revision: D6346241
fbshipit-source-id: 3fe6f20
Summary: This information is already available in the procedure name.
Reviewed By: jeremydubreil, jvillard
Differential Revision: D6119459
fbshipit-source-id: f07bfde
Summary:
First steps of a dispatcher for C++ functions/methods overloads.
For now only used on Inferbo C modeled functions so most of the features are still unused.
Reviewed By: jvillard
Differential Revision: D6336088
fbshipit-source-id: ebd5b6f
Summary:
...so I just removed it
+ renamed `loc` of type `Location.t` to `location` to differentiate from `Loc.t` values
Reviewed By: jvillard
Differential Revision: D6358413
fbshipit-source-id: 2d3eba9
Summary: The clang compiler introduces a materialized temporary expression which should be treated similarly to the Infer internal temporary variables.
Reviewed By: sblackshear
Differential Revision: D6331237
fbshipit-source-id: 81d8196
Summary:
We would previously skip any function that had one of these.
A no-op translation is sufficient to fix this issue (see new E2E test).
Reviewed By: mbouaziz
Differential Revision: D6317323
fbshipit-source-id: 0855bd8
Summary:
`infer capture -a checkers ...` would accidentally trigger the analysis phase.
This crashes the Buck flavors integration when used with `--reactive` because
.start never gets created in the infer-out-* subfolders of buck-out.
Reviewed By: dulmarod
Differential Revision: D6336072
fbshipit-source-id: af0ab5e
Summary:
Target patterns/aliases rarely contain only targets supporting infer flavor, so it makes sense to automatically filter kinds with handle in those cases.
No need for `$(buck query ...)` anymore in your infer commands!
Reviewed By: dulmarod
Differential Revision: D6335463
fbshipit-source-id: 16c8b70
Summary:
When refactoring `Buck.ml` I took the list of accepted kinds that was used for compilation database.
However `#infer-capture-all` flavor is not supported by `cxx_test` targets.
Reviewed By: dulmarod
Differential Revision: D6335543
fbshipit-source-id: db3a5f4
Summary:
To resole #797 this adds runtime option to select VFS for SQLite,
When infer runs on WSL this defaults to "unix-excl" (https://sqlite.org/vfs.html) and if VFS is specified, then WAL is not enabled (since WAL is non compatible with custom VFS - https://www.sqlite.org/wal.html).
Closes https://github.com/facebook/infer/pull/798
Reviewed By: jvillard
Differential Revision: D6335037
Pulled By: dulmarod
fbshipit-source-id: d9b9a58
Summary: Just changing ClangTrace to actually look at the different sanitizer kinds.
Reviewed By: jeremydubreil
Differential Revision: D6325086
fbshipit-source-id: 5da236d
Summary:
We need to use the procedure description of the callees for lazy dynamic dispatch and for the resolution of the lambda. We may also need this information in other analyses, e.g. for RacerD. This diff makes the procedure description of the callees as part of the summary.
The procedure description has been part of the summary for a while already without noticeable decrease in performance.
Reviewed By: mbouaziz
Differential Revision: D6322038
fbshipit-source-id: 84101cb
Summary: This does not seem to be used anymore. If we happen to need this, we should update the payload, not the attributes.
Reviewed By: jberdine
Differential Revision: D6321824
fbshipit-source-id: 5c19359
Summary: In a thread safety report we used the access path from the final sink. This diffs change the report to include the expanded access path from the initial sink.
Reviewed By: sblackshear
Differential Revision: D6297848
fbshipit-source-id: 2386063
Summary: Having a summary for a callee from the specs cache does not necessarily mean that Eradicate has been run on it. This diff looks at the Eradicate payload instead from the return of the on-demand analysis instead.
Reviewed By: sblackshear
Differential Revision: D6054376
fbshipit-source-id: c6eec35
Summary: In the translation from SIL to HIL we ignore the right-hand side expression if it consists of a single access path, e.g. unary operator. This diff preserves the right-hand side expression.
Reviewed By: sblackshear
Differential Revision: D6271814
fbshipit-source-id: c27e913
Summary:
Change ocamlformat installation procedure to use opam instead of
pinning.
Reformat all code with v0.2, which has a few improvements.
Reviewed By: jvillard
Differential Revision: D6292057
fbshipit-source-id: 759967f
Summary:
This diff adds a new way of executing blocks when they are passed as parameters to a method. So far we just skipped the block in this case.
Now we can execute it. Let's demonstrate with an example. Say we have
//foo has a block parameter that it executes in its body
foo (Block block) { block();}
// bar calls foo with a concrete block
bar() {
foo (^(){
self->x = 10;
});
};
Now, when we call the method foo with a concrete block, we create a copy of foo instantiated with the concrete block, which in itself is translated as a method with a made-up name.
The copy of foo will get a name that is foo extended with the name of the block parameter, the call to the block parameter will be replaced to a call to the concrete block, and the captured variables
of the concrete block (self in this case), will be added to the formals of the specialized method foo_block_name.
This is turned on at the moment for ObjC methods with ObjC blocks as parameters, and called with concrete blocks. Later on we can extend it to other types of methods, and to C++ lambdas, that are handled similarly to blocks.
Another extension is to check when the block has been called with nil instead of an actual block, and raise an error in that case.
After this diff, we can also model various methods and functions from the standard library that take blocks as parameters, and remove frontend hacks to deal with that.
Reviewed By: ddino
Differential Revision: D6260792
fbshipit-source-id: 0b6f22e
Summary: The checker should not report unitinialzed values on the throw branch.
Reviewed By: ddino
Differential Revision: D6267019
fbshipit-source-id: 05768f1
Summary:
When fuzzy-matching cpp names, allow to match only a prefix of
blacklist entries.
Reviewed By: da319
Differential Revision: D6233055
fbshipit-source-id: a3a4913
Summary: We were conflating reads/writes with container reads/writes that created false positives.
Reviewed By: sblackshear
Differential Revision: D6232768
fbshipit-source-id: 39159cb
Summary: Better error message for the direct dereference of nullable method without intermediate variable.
Reviewed By: sblackshear
Differential Revision: D6244494
fbshipit-source-id: 2ca2d22
Summary: This is a hack to removes most of the false positives of this checker in Objective C.
Reviewed By: sblackshear
Differential Revision: D6239914
fbshipit-source-id: 1cf05de
Summary:
Update plugin to take into account that some fields of VarDecl were unused by
infer. Also, use a boolean holding `hasExternalStorage` instead of comparing to
the fragile (and probably not entirely accurate) `"extern"` string.
update-submodule: facebook-clang-plugins
Reviewed By: mbouaziz
Differential Revision: D6231836
fbshipit-source-id: 3c0a75b
Summary:
This confuses the SIOF checker and causes false positives. This dummy deref is
generated for constructors of classes that are modeled as being pointer types
instead of the actual class in infer, typically for smart pointers. I do not
understand how this works.
The biabduction also analyses this code, so might now get confused itself.
Reviewed By: jberdine
Differential Revision: D6221817
fbshipit-source-id: 050c5a9
Summary:
The issue is with classes defining static data members:
```
$ cat foo.h
struct A {
static int foo;
};
$ cat foo.cpp
#include "foo.h"
int A::foo = 12;
int f() { return A::foo; // should see A::foo as defined in this translation unit
$ cat bar.cpp
#include "foo.h"
void g() { return A::foo; // should see A::foo defined externally
```
Previously, both foo.cpp and bar.cpp would see `A::foo` as defined within their
translation unit, because it comes from the header. This is wrong, and static
data members should be treated as extern unless they're defined in the same
file.
This doesn't change much except for frontend tests. SIOF FP fix in the next diff.
update-submodule: facebook-clang-plugins
Reviewed By: da319
Differential Revision: D6221744
fbshipit-source-id: bef88fd
Summary: This only works for Java at the moment but we can re-organise the code later to add the Objective C equivalent of these assertion methods.
Reviewed By: mbouaziz
Differential Revision: D6230588
fbshipit-source-id: 46ee98e
Summary:
When C++ functions are translated to SIL procedures, their type is C rather then C++. In RacerD, we want to treat C++ functions the same as C++ methods.
Added a function to check if the procedure is Objc/Objc++/C/C++.
Reviewed By: sblackshear
Differential Revision: D6209523
fbshipit-source-id: 293f938
Summary:
Seems it should have been done there all along.
The analyzer does not currently understand the implementation of
atomicity in folly::AtomicStruct.
The analyzer does not currently understand when std::atomic operations
are are used correctly versus incorrectly.
The analyzer does not currently understand that the representation of
folly::ThreadLocal is, ah, thread-local, leading to false alarms.
The analyzer does not currently understand the control flow /
scheduling constraints imposed by the implementation of Future.
It seems that the implementation of folly::Optional is more C++
template magic than the analyzer can currently understand.
The model of std::vector contains bogus memory accesses, leading to
false alarms.
Reviewed By: sblackshear
Differential Revision: D6226199
fbshipit-source-id: 8cb083b
Summary:
Destructors usually do not race with other methods.
We do not want to analyze or report on destructors.
Reviewed By: sblackshear
Differential Revision: D6222145
fbshipit-source-id: 5266622
Summary:
:
As we want to model many C++ methods, using a lot of matchers with `if / else if` will be tiring.
This diff introduces a dispatcher which is a nicer way to write the same thing.
No new model for now, just a refactoring.
Ideally we'd need a parser generator for C++ names...
Reviewed By: jvillard
Differential Revision: D6209234
fbshipit-source-id: 49fae5e
Summary: The checker should not report nullable violations on repeated calls
Reviewed By: sblackshear
Differential Revision: D6195471
fbshipit-source-id: 16ff76d
Summary: The Java bytecode does not contain information about the location of abstract of interface methods. Before this diff, the analysis trace was tuncated and the file where the abstract or interface method was not included in the trace, which makes it harder to understand the Infer report, especially when the method is on a generated file that is not checked in the repository.
Reviewed By: sblackshear
Differential Revision: D6223612
fbshipit-source-id: c80c6f2
Summary: A source can belong to more than one target. In this case, we should keep only one of the report.
Reviewed By: sblackshear
Differential Revision: D6200058
fbshipit-source-id: 4eced42
Summary:
The SIOF checker relies on the header models to detect whether `<iostream>` has
been included in source files.
Reviewed By: mbouaziz
Differential Revision: D6209904
fbshipit-source-id: a48855b
Summary: More general version of the fix in D6138749. This diff moves RacerD's lock modeling into a separate module and uses the module in the HIL translation to check when a function has lock/unlock semantics.
Reviewed By: jberdine, da319
Differential Revision: D6191886
fbshipit-source-id: 6e1fdc3
Summary:
This diff takes the first step toward a more general filtering
system. This step is concerned only with filtering at the reporting
stage, filtering for the capture and analysis stages is left for
later.
This diff adds a new command line / config option
```
--filter-report +string
Specify a filter for issues to report. If multiple filters are
specified, they are applied in the order in which they are
specified. Each filter is applied to each issue detected, and only
issues which are accepted by all filters are reported. Each filter
is of the form:
`<issue_type_regex>:<filename_regex>:<reason_string>`. The first
two components are OCaml Str regular expressions, with an optional
`!` character prefix. If a regex has a `!` prefix, the polarity is
inverted, and the filter becomes a "blacklist" instead of a
"whitelist". Each filter is interpreted as an implication: an issue
matches if it does not match the `issue_type_regex` or if it does
match the `filename_regex`. The filenames that are tested by the
regex are relative to the `--project-root` directory. The
`<reason_string>` is a non-empty string used to explain why the
issue was filtered.
See also infer-report(1) and infer-run(1).
```
Reviewed By: jvillard
Differential Revision: D6182486
fbshipit-source-id: 9d3922b
Summary: Functions that do not belong to a class or a struct are translated to c-style functions even in the context of cpp. We need to add ownership to locals for c-style functions too.
Reviewed By: sblackshear
Differential Revision: D6196882
fbshipit-source-id: 715f129
Summary:
vector::data returns a pointer to the first value of the vector.
- The size of the (array) pointer should be the same with the vector.
- The pointer should point to the same abstract value with the vector.
Reviewed By: mbouaziz
Differential Revision: D6196592
fbshipit-source-id: cc17096
Summary: `std::unique_lock` constructor allows to create a unique lock without locking the mutex. `std::unique_lock::try_lock` returns true if mutex has been acquired successfully, and false otherwise. It could be that an exception is being thrown while trying to acquire mutex, which is not modeled.
Reviewed By: jberdine
Differential Revision: D6185568
fbshipit-source-id: 192bf10
Summary:
The concurrency analyzer often does not understand object lifetimes
well enough to realize that destructors are usually not called in
parallel with any other methods. This leads to false alarms. This diff
suppresses these by simply skipping destructors in the concurrency
analysis.
Reviewed By: sblackshear
Differential Revision: D6182646
fbshipit-source-id: e9d1cac
Summary:
The clang frontend translates static locals incorrectly, in the sense
that the initializer is executed many times instead of once. This
leads to false alarms in the concurrency analysis. This diff
suppresses these by ignoring accesses to static locals.
Reviewed By: sblackshear
Differential Revision: D6182644
fbshipit-source-id: d8ca4c0
Summary:
Code often uses std::unique_lock::owns_lock to test if a deferred lock
using the 2-arg std::unique_lock constructor actually acquired the
lock.
Reviewed By: sblackshear
Differential Revision: D6181631
fbshipit-source-id: 11e9df2
Summary:
Use a distinct issue type for the Java and C++ concurrency analyses,
as the properties they are checking are significantly different.
Reviewed By: sblackshear
Differential Revision: D6151682
fbshipit-source-id: 00e00eb
Summary:
In a summary, you never want to see a trace where non-footprint sources flow to a sink.
Such a trace is useless because nothing the caller does can make more data flow into that sink.
Reviewed By: jeremydubreil
Differential Revision: D5779983
fbshipit-source-id: d06778a
Summary:
Due to limitations in our Buck integration, the thread-safety analysis cannot create a trace that bottoms out in a Buck target that is not a direct dependency of the current target.
These truncated traces are confusing and tough to act on.
Until we can address these limitations, let's avoid reporting on truncated traces.
Reviewed By: jeremydubreil
Differential Revision: D5969840
fbshipit-source-id: 877b9de
Summary:
Relative paths in jbuilder + `S **` seem to be a losing combo. Spell out the directories instead.
This was obtained via letting jbuilder generate .merlin, then curating it by hand.
Reviewed By: jberdine
Differential Revision: D6159600
fbshipit-source-id: 7d799bb
Summary:
:
Make both buck capture and compilation database handle buck command line arguments and invoke buck query the same way.
Plus allow:
- target patterns `//some/dir:` and `//some/dir/...`. However since `//some/dir:#flavor` and `//some/dir/...#flavor` are not supported, they need to be expanded before adding the infer flavor.
- target aliases (defined in `.buckconfig`)
- shortcuts `//some/dir` rewritten to `//some/dir:dir`
- relative path `some/dir:name` rewritten to `//some/dir:name`
Reviewed By: jvillard
Differential Revision: D5321087
fbshipit-source-id: 48876d4
Summary: These can make the compilation fail, so don't use them unless we really need to.
Reviewed By: mbouaziz
Differential Revision: D6147574
fbshipit-source-id: ab2c3fa
Summary:
If you write
```
boolean readUnderLockOk() {
synchronized (mLock) {
return mField;
}
}
```
it will be turned into
```
lock()
irvar0 = mField
unlock()
return irvar0
```
in the bytecode. Since HIL eliminates reads/writes to temporaries, it will make the above code appear to perform a read of `mField` outside of the lock.
This diff fixes the problem by forcing HIL to perform all pending reads/writes before you exit a critical section.
Reviewed By: jberdine
Differential Revision: D6138749
fbshipit-source-id: e8ad9a0
Summary: In HIL, allow deref'ing a magic address like `0xdeadbeef` for debugging purposes. Previously, we would crash on code like this.
Reviewed By: mbouaziz
Differential Revision: D6143802
fbshipit-source-id: 4151924
Summary:
Linters are now considered a "checker", like backend checkers. This makes, eg,
`--racerd-only` disable the linters, which is more intuitive.
We can now express `-a linters` and `--clang-frontend-action` in terms of these
two new options. For instance, `-a linters --clang-frontend-action lint` is the
same as `--linters-only --no-capture`.
This is another step in the direction of getting rid of `--analyzers`.
Reviewed By: dulmarod
Differential Revision: D6147387
fbshipit-source-id: 53622b2
Summary: A stepping stone to have descriptive issue types for each kind of flow rather that lumping everything into `QUANDARY_TAINT_ERROR`.
Reviewed By: mbouaziz
Differential Revision: D6126690
fbshipit-source-id: a7230c0
Summary: This check is deprecated and will be replaced by a dedicated checker to detect unitialized values.
Reviewed By: mbouaziz
Differential Revision: D6133108
fbshipit-source-id: 1c0e9ac
Summary: Previously, this would incorrectly classify types like `map<std::string, int>` as a buffer
Reviewed By: mbouaziz
Differential Revision: D6125530
fbshipit-source-id: c8564de
Summary:
Before this change, analyses using HIL needed to pass `IdAcessPathMapDomain.empty` to abstract interpreter, and would get back the map as part of the post.
This is a confusing API and was a pain point for Dino in trying to use HIL.
This diff adds a HIL wrapper around the abstract interpreter that hides these details.
It replaces `LowerHIL.makeDefault` as the new "simplest possible way" to use HIL.
Reviewed By: jberdine
Differential Revision: D6125597
fbshipit-source-id: 560856b
Summary:
This is in the spec for clang compilation databases.
Also improves error messages when we fail to parse the compilation database.
closes#771
Reviewed By: dulmarod
Differential Revision: D6123832
fbshipit-source-id: 070f70f
Summary: Saving the list of bugs in a set removes the ordering. Also, there should be no need to remove the duplicated warnings at this level.
Reviewed By: sblackshear
Differential Revision: D6060554
fbshipit-source-id: a78d35d
Summary:
Looked at some problematic summaries and am noticing some common patterns.
Adding some dynamic checks to be run in debug mode in order to make sure my fixes for these patterns are real.
Reviewed By: jeremydubreil
Differential Revision: D5779593
fbshipit-source-id: 9de6497
Summary:
The options passed via `--Xbuck` are usually meant for `buck build` but we also
use them for `buck targets`. It's hard to know which options to take into
account for which Buck subcommand, so just filter out known-incompatible ones.
Reviewed By: dulmarod
Differential Revision: D6123459
fbshipit-source-id: 976b978
Summary:
Install ocamlformat from github as part of `make devsetup`, and use it
for formatting OCaml (and jbuild) code.
Reviewed By: jvillard
Differential Revision: D6092464
fbshipit-source-id: 4ba0845
Summary: Sinks weren't being printed when passthroughs are empty (which, for now, is always). Oops!
Reviewed By: jvillard
Differential Revision: D6110164
fbshipit-source-id: 4488ab0
Summary: This will make it easier to generalize the checker to handling uninitialized struct fields.
Reviewed By: ddino
Differential Revision: D6099484
fbshipit-source-id: b9c534b
Summary:
Buck reads the version on stderr or, very recently, from either stdout or stderr.
This makes infer output the version of stderr when called from Buck or invoked as javac, and on stdout otherwise.
Reviewed By: martinoluca
Differential Revision: D6098392
fbshipit-source-id: 23f1d5a
Summary: This makes `--biabduction-blacklist-path-regex` and others work as expected.
Reviewed By: mbouaziz
Differential Revision: D6088625
fbshipit-source-id: 8f1daa3
Summary:
This is a better default than running the biabduction analysis only, now that
we have several mature checkers.
Reviewed By: jeremydubreil
Differential Revision: D6051186
fbshipit-source-id: 04ac0c6
Summary: In preparation for making `-a checkers` the default (when no analyzer is specified), let's test `-a checkers` by default.
Reviewed By: mbouaziz
Differential Revision: D6051177
fbshipit-source-id: d8ef611
Summary:
Refactor `RegisterCheckers` to give a record type to checkers instead of a tuple type.
Print active checkers with their per-language information.
Improve the manual entries slightly.
Reviewed By: sblackshear
Differential Revision: D6051167
fbshipit-source-id: 90bcb61
Summary:
Whenever we see a use of a lock, infer that the current method can run in a multithreaded context. But only report when there's a write under a lock that can be read or written without synchronization elsewhere.
For now, we only infer this based on the direct usage of a lock; we don't assume a caller runs in a multithreaded context just because its (transitive) callee can.
We can work on that trickier case later, and we can work on smarter inference that takes reads under sync into account. But for now, warning on unprotected writes of reads that occur under sync appears to be too noisy.
Reviewed By: jberdine
Differential Revision: D5918801
fbshipit-source-id: 2450cf2
Summary: This commit adds unsigned symbol for preciser analysis results with less number of uses of min/max operators.
Reviewed By: mbouaziz
Differential Revision: D6040437
fbshipit-source-id: 999ca4c
Summary:
This is useful if some command behaves like one infer knows how to integrate with. For instance:
```
infer --force-integration clang -- clang-3.8 -c examples/hello.c
```
Reviewed By: jeremydubreil, mbouaziz
Differential Revision: D6051589
fbshipit-source-id: dd693b0
Summary:
This allows us to get rid of code that copied source files individually. I
didn't migrate the various flags that could be included as it doesn't look like
that's possible yet (they depend on the context and on some configuration
options).
Reviewed By: jberdine
Differential Revision: D6051825
fbshipit-source-id: c28dd37
Summary:
This will allow most of the checkers, except the bi-abduction, to skip the analysis on the specialized clone of the methods used to handle dynamic dispatch. Doing this, we can run the bi-abduction analysis using:
infer -a checkers --biabduction
without risk of conflicts on the resolution of dynamic dispatch.
Reviewed By: sblackshear
Differential Revision: D6052347
fbshipit-source-id: 0c75bf3
Summary: This removes cases of duplicated warnings when the dynamic dispatch handling specializes a method Infer already reported on.
Reviewed By: sblackshear
Differential Revision: D6060337
fbshipit-source-id: dbefeca
Summary:
It looks like the old code for expanding access paths assumed that `FormalMap.get_formals_indexes` assumed the returned list tuples would be sorted by index, but it's actually sorted by var name.
As a consequence, formals might be expanded into the wrong actuals.
This diff fixes the problem by not relying on `get_formals_indexes`.
Reviewed By: jberdine
Differential Revision: D6056365
fbshipit-source-id: 09f3208
Summary: The order of the elements in the list maters since the function `string_to_analyzer` will return the fist element found in the list. Inverting the `"biabduction"` and `"infer"` entries in the list allows D6051146 to have no functional implications.
Reviewed By: sblackshear
Differential Revision: D6055840
fbshipit-source-id: 6cf5ac2
Summary:
This was a crutch from the days before ownership analysis.
We shouldn't need it anymore, and it was actually causing FP's because we were skipping analysis of `ImmutableList.builder()` and not understanding that the return value is owned.
Reviewed By: jeremydubreil
Differential Revision: D6035631
fbshipit-source-id: afa0ade
Summary:
One day `-a infer` will alias `-a checkers` so for now create another, more
explicit analyzer name that can be used to migrate progressively. For instance,
after this commit what should continue to use `-a biabduction` can change to
that, so that when `-a infer` becomes an alias to `-a checkers` it can keep
working.
Reviewed By: jeremydubreil
Differential Revision: D6051146
fbshipit-source-id: 1ef4c34
Summary:
This generates `--resource-leak-only` automatically, and make the other
checkers' `-only` option work as expected with respect to `--resource-leak` too
(eg, `--resource-leak --biabduction-only` disables resource leak).
Reviewed By: jeremydubreil
Differential Revision: D6051134
fbshipit-source-id: 2d4a2ba
Summary:
1. Mark some Makefile targets as depending on `MAKEFILE_LIST` so they get rebuilt on Makefile changes
2. Do not show boolean options with no documentation in the man pages (like we do for other option types).
3. Default to Lazy dynamic dispatch for the checkers.
4. In the tests, use `--<checker>-only` instead of relying on `--no-default-checkers`
5. `--no-filtering` is redundant if `--debug-exceptions` is passed
Reviewed By: jeremydubreil
Differential Revision: D6030578
fbshipit-source-id: 3320f0a
Summary: We will then be able to merge the tests for the other checkers without affecting these lab tests
Reviewed By: jvillard
Differential Revision: D6039433
fbshipit-source-id: e575ce9
Summary:
Another step toward running the biabduction analysis as a checker.
Depends on D6038210
Reviewed By: jvillard
Differential Revision: D6038682
fbshipit-source-id: fed45bf
Summary:
The previous version of the code was trying to lookup from disk the procedure description of the procedure to analyze, which was in fact already loaded in memory.
This diff fixes one of the issues preventing the bi-abduction to run as a checker when using the lazy dynamic dispatch algorithm.
Reviewed By: sblackshear
Differential Revision: D6038210
fbshipit-source-id: 10a98ee
Summary:
9c7fc65 introduced a large performance regression, this diff eliminates it and a bit more.
Instead of constructing the quotiented access list map in a two-step process of first constructing a map of all accesses and then quotienting it, the quotiented map is constructed directly by using a coarser comparison function on keys. Partitioning the access map O(number of access paths) times, using an apparently expensive partition predicate, seems to be causing trouble based on rough profile data.
Reviewed By: da319
Differential Revision: D6005262
fbshipit-source-id: 077846c
Summary: Stack-allocated variables cannot be raced on in cpp as every thread has its own stack. At the beginning of the analysis we add ownership to the local variables.
Reviewed By: jberdine
Differential Revision: D6020506
fbshipit-source-id: 0a90a97
Summary: Now that we report write-write races involving more than one write, we need to improve the traces accordingly.
Reviewed By: jberdine
Differential Revision: D6026845
fbshipit-source-id: b1366dd
Summary:
This is to avoid getting BUSY from sqlite when the machine is busy (not
necessarily busy because of infer).
Reviewed By: jberdine
Differential Revision: D6020022
fbshipit-source-id: ca0f913
Summary:
Next step to issue deduplication: do not keep safety conditions that are subsumed by others.
Only do it if they do not have infinite bound: replacing `0 < size` by `1 < size` is ok, but replacing it by `+oo < size` is not because it looks much more like a lack of precision.
Reviewed By: skcho
Differential Revision: D5978455
fbshipit-source-id: acc2384
Summary:
A specific type of alias is added for the vector::empty() result and it is used at pruning.
Now, there are two types of aliases:
- "simple" alias: x=y
- "empty" alias: x=v.empty() and y=v.size
So, if x!=0, y is pruned by (y=0). Otherwise, i.e., x==0, y is pruned by (y>=1).
Reviewed By: mbouaziz
Differential Revision: D6004968
fbshipit-source-id: bb8d50d
Summary:
`pp_instr_list` was not tailrec causing a stack overflow on big code.
Also simplified a few things
Reviewed By: jvillard
Differential Revision: D5995451
fbshipit-source-id: 40a4911
Summary: The may alias analysis relation that the thread safety analysis uses is very specific to Java and causes many false alarms for C++ code. This diff disables it when analyzing C++ code. Improving it to avoid false negatives is left for later.
Reviewed By: sblackshear
Differential Revision: D5974182
fbshipit-source-id: 9c7fc65
Summary:
The analyzer currently does not understand the control flow of
Singletons, which leads to false alarms. This diff is an unsound hack
that simply ignores any read or write accesses made when computing the
value of a singleton.
Reviewed By: sblackshear
Differential Revision: D5979639
fbshipit-source-id: 34caecb
Summary:
Model folly::SharedMutex lock and unlock operations, some
apache::thrift::concurrency::ReadWriteMutex operations, some
folly::RWSpinLock operations, and folly::MicroSpinLock operations.
Reviewed By: sblackshear
Differential Revision: D5974225
fbshipit-source-id: 19e2816
Summary:
The biabduction backend can raise exceptions that will be caught when triggered
from within the biabduction backend itself (eg, `analyze_procedure` called from
Symexec as a result of an ondemand analysis, because Symexec will catch these),
but not caught when called as the result of an ondemand analysis emanating from
another analyzer (eg ThreadSafety).
Make the biabduction more self-contained by wrapping the analysis of a
procedure inside a `try/with` with similar properties as the one of Symexec.
Reviewed By: jeremydubreil
Differential Revision: D5986335
fbshipit-source-id: 36a5d32
Summary:
Attempting to translate these will not go well as the declaration still depends
on some template arguments. Added a test that was previously crashing the
frontend.
Also extend the catching of "Unimplemented" and other errors to `translate_one_decl` as it was useful to debug this issue. In particular, reraise all exceptions and log some additional context when doing so.
update-submodule: facebook-clang-plugins
Reviewed By: mbouaziz
Differential Revision: D5976357
fbshipit-source-id: fca8e38
Summary:
Previously, annotating something ThreadSafe meant "check that it is safe to run all of this procedure's methods in parallel with each other" (including self-parallelization).
This makes sense, but it means that if the user writes no annotations, we do no checking.
I'm moving toward a model of inferring when an access might happen on a thread that can run concurrently with other threads, then automatically checking that it is thread-safe w.r.t to all other accesses to the same memory (on or off the current thread thread).
This will let us report even when there are no `ThreadSafe` annotations.
Any method that is known to run on a new thread (e.g., `Runnable.run`) will be modeled as running on a thread that can run in parallel with other threads, and so will any method that is `synchronized` or acquires a lock.
In this setup, adding `ThreadSafe` to a method just means: "assume that the current method can run in parallel with any thread, including another thread that includes a different invocation of the same method (a self race) unless you see evidence to the contrary" (e.g., calling `assertMainThread` or annotating with `UiThread`).
The key step in this diff is changing the threads domain to abstract *what threads the current thread may run in parallel with* rather than *what the current thread* is. This makes things much simpler.
Reviewed By: jberdine
Differential Revision: D5895242
fbshipit-source-id: 2e23d1e
Summary:
Indicate if read or write is protected, and do not print only the
field but also the object involved in the race.
Reviewed By: sblackshear
Differential Revision: D5974250
fbshipit-source-id: 351a576
Summary:
Expanding traces currently works in the following way:
Given a `TraceElem.Kind` `k` we want to report in `foo`, we look for a callee `C` of `foo` that has a `TraceElem.Kind` equal to `k` in its summary, grab the summary for `C`, then repeat until we bottom out.
This isn't very flexible: it insists on equality between `TraceElem.Kind`'s as the criteria for expanding a trace.
This diff introduces a new `matches` function for deciding when to expand a trace from a caller into a callee.
Clients that don't want strict equality can implement a fuzzier kind of equality inside this function.
I've gone ahead and done this for the trace elemes of thread-safety.
In the near future, equivalent access paths won't always compare equal from caller to callee, so we want to match their suffixes instead.
Reviewed By: jvillard
Differential Revision: D5914118
fbshipit-source-id: 233c603
Summary:
Move Inferbo safety conditions to their own file.
Split the old `Condition.t` to a condition together with a trace.
This will ease having: different kind of condition and several traces for the same condition (see following diff)
Reviewed By: jvillard
Differential Revision: D5942030
fbshipit-source-id: d74a612
Summary: Not using this for now, and it seems good to simplify the complex domain as much as we can.
Reviewed By: jberdine
Differential Revision: D5970233
fbshipit-source-id: a451503
Summary:
This is due to the changes in `facebook-clang-plugins` where objc_object_type_info now has a `field_prefix` set to `ooti_`
See 5f2042abe6 for the changes made to `facebook-clang-plugins`
update-submodule: facebook-clang-plugins
Reviewed By: jvillard
Differential Revision: D5963064
fbshipit-source-id: 9705774
Summary:
Running `infer report foo.specs` would overwrite report.json with an empty json.
Only recompute report.json when running `infer-analyze` or `infer-run`.
Reviewed By: mbouaziz
Differential Revision: D5963500
fbshipit-source-id: e579c6f
Summary:
Use a monotonic time source instead.
Also, sleep between retries in the Serialization code.
Reviewed By: jberdine
Differential Revision: D5941697
fbshipit-source-id: 05efbe1
Summary:
Despite what the sqlite manual says, it looks like it's possible for sqlite not
to remove the -shm and -wal files after a successful termination.
Close the database and clean them up in Buck mode so that they do not perturb
the cache.
Reviewed By: martinoluca
Differential Revision: D5953967
fbshipit-source-id: 9068b01
Summary:
Inject a marker using a global variable in <iostream>, and whitelist it so that
the frontend translates it.
Use the marker in the SIOF checker to tell whether a file includes <iostream>.
If so, start the analysis of its methods assuming that the standard streams are
initialised.
Reviewed By: sblackshear
Differential Revision: D5941343
fbshipit-source-id: 3388d55
Summary:
The previous domain for SIOF was duplicating some work with the generic Trace
domain, and basically was a bit confused and confusing. A sink was a set of
global accesses, and a state contains a set of sinks. Then the checker has to
needlessly jump through hoops to normalize this set of sets of accesses into a
set of accesses.
The new domain has one sink = one access, as suggested by sblackshear. This simplifies
a few things, and makes the dedup logic much easier: just grab the first report
of the list of reports for a function.
We only report on the fake procedures generated to initialise a global, and the
filtering means that we keep only one report per global.
Reviewed By: sblackshear
Differential Revision: D5932138
fbshipit-source-id: acb7285
Summary:
Wrap ANSITerminal so that it doesn't apply control codes unless they can be
interpreted, and so that clients are unable to call ANSITerminal directly.
Reviewed By: mbouaziz
Differential Revision: D5953733
fbshipit-source-id: 6b3602a
Summary:
Bottom bounds do not make sense (what is the meaning of `[_|_; 1]`?), let's get rid of them.
`Bot` was useful for substitution though, with a special meaning, use `bottom_lifted` for that case.
Reviewed By: skcho
Differential Revision: D5941796
fbshipit-source-id: 5778255
Summary: A bottom interval in a safety condition doesn't make sense. Let's not allow it at all.
Reviewed By: skcho
Differential Revision: D5941552
fbshipit-source-id: 6bd2a65
Summary: If we know for sure we won't need to store an attribute in the DB, there's no need to compute its marshalled value.
Reviewed By: jberdine
Differential Revision: D5891050
fbshipit-source-id: cf4534e
Summary:
This adds more structure to the SQL schema backing attributes. With that, we
can transfer the logic for updating attributes in SQLite, instead of doing
optimistic concurrency in the client.
Reviewed By: jberdine
Differential Revision: D5891038
fbshipit-source-id: 6577ba2
Summary:
- use a similar key as for specs in the attributes table
- cache blob computations
- this improves memory usage a lot
Reviewed By: mbouaziz
Differential Revision: D5824177
fbshipit-source-id: c318577
Summary:
Preparing statements allocates memory and is generally best done once and for all.
We need to re-prepare statements when opening a new DB connexion (after each fork).
Reviewed By: mbouaziz
Differential Revision: D5824157
fbshipit-source-id: 4d239ac
Summary: Not sure if useful but seems sensible. It disappears at the top of the stack when we do merging in SQL.
Reviewed By: mbouaziz
Differential Revision: D5824131
fbshipit-source-id: fd64752
Summary: Use `pp` functions until it needs to be turned into a string.
Reviewed By: jvillard
Differential Revision: D5941473
fbshipit-source-id: 87ca9df
Summary:
Use an SQLite database to store proc attributes, instead of files on disk.
Wrap SQLite operations in two layers:
1. `SqliteUtils` provides helper functions to make sure DB operations succeed
2. `KeyValue` provides a functor to expose a simple and type-safe key/value store backed by the SQLite DB.
Reviewed By: jberdine
Differential Revision: D5640053
fbshipit-source-id: 31050e5
Summary:
Its value is unused in Infer and is constantly emitted as None from facebook-clang-plugins, so it was also removed from facebook-clang-plugins (a96c39601f)
update-submodule: facebook-clang-plugins
Reviewed By: jvillard
Differential Revision: D5940900
fbshipit-source-id: e7fd6ae
Summary: Remove functions that are redundant with Base.List functions on sorted lists.
Reviewed By: mbouaziz
Differential Revision: D5931792
fbshipit-source-id: caec210
Summary: The case of closures was not considered for the convertion of SIL instructions into HIL instructions
Reviewed By: sblackshear
Differential Revision: D5929675
fbshipit-source-id: bb6920a
Summary: These folders are left empty when linting, so better not create them.
Reviewed By: jvillard
Differential Revision: D5921418
fbshipit-source-id: f6efc4f
Summary: The tests are slower when running in debug mode, and it creates a lot of html outputs
Reviewed By: sblackshear
Differential Revision: D5916511
fbshipit-source-id: 07c90b7
Summary:
This diff does two things:
# Infer no longer add the contrains that the return value of a skip function is never null. This was leading to false negatives and is not necessary as those return value are treated angelically
# Infer now support `Nonnull` on the return value of skip functions.
Reviewed By: jberdine, sblackshear
Differential Revision: D5840324
fbshipit-source-id: bbd8d82
Summary:
`reraise` was error-prone when one forgot to save the backtrace between where the exception is caught and where it is reraised.
If any exception was raised (even caught) in between, the printed backtrace would be the one of the last exception thrown and it would be very confusing.
This diff kills `reraise` and introduces `reraise_after exn ~f` and `reraise_if exn ~f` to be used right after catching the exception.
Also turned some of them to the common pattern `try_finally ~f ~finally`.
Reviewed By: jvillard
Differential Revision: D5911244
fbshipit-source-id: 9883d1e
Summary: Example of combination between annotating fields with nullable and the biabduction analysis in Objective C
Reviewed By: dulmarod
Differential Revision: D5906016
fbshipit-source-id: b95c6e0
Summary:
The interval bound of the abstract domain is extended.
`[min|max](int, symbol)` => `int [+|-] [min|max](int, symbol)`
As a result, `vector::empty` can be modelled to return a more precise value: `1 - min(1, size)`.
Reviewed By: mbouaziz
Differential Revision: D5899404
fbshipit-source-id: c8c3f49
Summary:
We take it into account to not report bugs inside the available block. This requires a plugin change.
update-submodule: facebook-clang-plugins
Reviewed By: jvillard
Differential Revision: D5891511
fbshipit-source-id: 21a02ad
Summary:
With this change and the previous facebook-clang-plugins change, infer no
longer exhausts the biniou buffer when reading the serialized C++ AST.
update-submodule: facebook-clang-plugins
Reviewed By: mbouaziz
Differential Revision: D5891081
fbshipit-source-id: cf48eac
Summary:
These get way too big in C++, and we only use the very first word of them, to
tell apart class from struct from union... so sad, very bad.
update-submodule: facebook-clang-plugins
Reviewed By: mbouaziz
Differential Revision: D5890594
fbshipit-source-id: 49e6284
Summary:
The only language types we have are Java/Clang/Python. The unit of analysis is a source file, and you can't write a source file that mixes two or more of these languages (to the best of my knowledge).
This diff simplifies using the assumption that all procedures in a file are written in the same language.
Reviewed By: jeremydubreil
Differential Revision: D5886942
fbshipit-source-id: 88c3759
Summary:
The only language types we have are Java/Clang/Python. The unit of analysis is a source file, and you can't write a source file that mixes two or more of these languages (to the best of my knowledge).
This diff simplifies using the assumption that all procedures in a file are written in the same language.
Reviewed By: jeremydubreil
Differential Revision: D5886942
fbshipit-source-id: 8555a16
Summary: Only Eradicate uses this, no need to create it for every checker.
Reviewed By: jeremydubreil
Differential Revision: D5886775
fbshipit-source-id: 7242437
Summary:
A Java cluster checker currently defines a "cluster" as all of the procedures in the same class.
But the cluster checker actually knows about all the procedures defined in the same source file.
In some checkers (such as thread-safety), we want to aggregate results across classes in the same file, not just methods in the same class.
This refactoring leaves the behavior the same for now, but will make it easier to do this in the near future.
Reviewed By: jeremydubreil
Differential Revision: D5885896
fbshipit-source-id: 0815fca
Summary:
Calling functions that raise exceptions (even if they get caught) may smudge
the backtraces we get from OCaml. We need to record the original backtrace
*before* calling such fuctions on the path between catching an exception and
reraising it.
Also change the heptuple returned by `Exceptions.recognize_exception` into a
record type, and make that function not raise when classifying exceptions.
Reviewed By: jberdine
Differential Revision: D5882934
fbshipit-source-id: 8e99fe8
Summary: Infer can then detect the resource leak when resources are stored into a container
Reviewed By: sblackshear
Differential Revision: D5887702
fbshipit-source-id: 6106cfb
Summary: The point of the tracing mode is to compute all the possible path leading to an error state. However, within a method, many of those paths are not feasibile in practice. This leads to many false alarms for the resource leak analysis.
Reviewed By: sblackshear
Differential Revision: D5888695
fbshipit-source-id: 2dbc57b
Summary: Model the `remove(...)` and the `clear()` method of `HashMap`.
Reviewed By: sblackshear
Differential Revision: D5887674
fbshipit-source-id: c3f40ee
Summary: Handling the utility functions for asserting that we're on background thread.
Reviewed By: jberdine
Differential Revision: D5863435
fbshipit-source-id: 3ad95b5
Summary:
Previously, we just tracked a boolean representing whether we were possibly on the main thread (true) or definitely not on the main thread (false).
In order to start supporting `Thread.start`, `Runnable.run`, etc., we'll need something more expressive.
This diff introduces a lattice:
```
Any
/ \
Main Background
\ /
Unknown
```
as the new threads domain. The initial value is `Unknown`, and we introduce `Main` in situations where we would have introduced `true` before.
This (mostly) preserves behavior: the main difference is that before code like
```
if (*) {
assertMainThread()
} else {
x.f = ...
}
```
would have recorded that the access to `x.f` was on the main thread, whereas now we'll say that it's on an unknown thread.
Reviewed By: peterogithub
Differential Revision: D5860256
fbshipit-source-id: efee330
Summary:
It's useful to be able to disable de-duplication on the command line with `--no-filtering`.
Gate de-duplication with `Config.filtering` and move the de-duplication tests to a new directory under the build systems tests.
Reviewed By: jeremydubreil
Differential Revision: D5865329
fbshipit-source-id: 5094f5b
Summary:
Since D5381239, infer is careful not to delete directories that do not "look
like" results directories on startup, in case the user passed, eg, `-o /`.
In our repo, lots of results dir are created by build/test of infer, and when
the version of infer changes and the expected contents of results directories
change then it might start refusing to delete the results directories created
with another version of infer.
Add an option to force infer to delete the results directory no matter how
dodgy it looks, and use it in our repo by adding the option in every
.inferconfig.
Reviewed By: mbouaziz
Differential Revision: D5870984
fbshipit-source-id: 09412de
Summary: Will then be easier to understand if some changes in the test results is legit or not.
Reviewed By: sblackshear
Differential Revision: D5863961
fbshipit-source-id: 7eb3f33
Summary: Parmap delivers a better scheduling, which works as a pipeline, as opposed to what existed before, which schedules processes in batches.
Reviewed By: jvillard
Differential Revision: D5678661
fbshipit-source-id: a632c71
Summary:
Suggesting to add `_Nullable` on the fields checked for, or assigned to, `nullptr` will allow the biabduction analysis to report null dereferences that are related to the lifetime of objects.
Depends on D5832147
Reviewed By: sblackshear
Differential Revision: D5836538
fbshipit-source-id: c1b8e48
Summary: The prune nodes where translated as `prune (expr = false)` and `prune ( expr != false)`. This case is a bit tricky to deconstruct in HIL. This diff translates the prune instructions as just `prune !expr` for the true branch and `prune expr` for the false branch.
Reviewed By: dulmarod
Differential Revision: D5832147
fbshipit-source-id: 2c3502d
Summary:
We need to make sure that destructors of virtual base classes are called only once. Similarly to what clang does, we have two destructors for a class: a destructor wrapper and an inner destructor.
Destructor wrapper is called from outside, i.e., when variables go out of scope or when destructors of fields are being called.
Destructor wrappers have calls to inner destructors of all virtual base classes in the inheritance as their bodies.
Inner destructors have destructor bodies and calls to destructor wrappers of fields and inner destructors of non-virtual base classes.
Reviewed By: dulmarod
Differential Revision: D5834555
fbshipit-source-id: 51db238
Summary:
The "placement new" operator `new (e) T` constructs a `T` in the pre-allocated memory address `e`.
We weren't translating the `e` part, which was leading to false positives in the dead store analysis.
Reviewed By: dulmarod
Differential Revision: D5814191
fbshipit-source-id: 05c6fa9
Summary:
With flambda (`-O3`), compilation time is ~5x slower, but the backend is ~25% faster!
To mitigate the atrocious compilation times, introduce a new `opt` build mode in the jbuilder files.
- build in "opt" mode by default from the toplevel (so that install scripts and external users get the fastest infer by default), in "default" mode by default from infer/src (since the latter is only called directly by infer devs, for faster builds)
- `make byte` is as fast as before in any mode
- `make test` will build "opt" by default, which is very slow. Solution for testing (or building the models) locally: `make BUILD_MODE=default test`.
- You can even change the default locally with `export BUILD_MODE=default`.
The benchmarks are to be taken with a sizable pinch of salt because I ran them only once and other stuff could be running in the background. That said, the perf win is consistent across all projects, with 15-20% win in wallclock time and around 25% win in total CPU time, ~9% win in sys time, and ~25% fewer minor allocations, and ~5-10% fewer overall allocations. This is only for the backend; the capture is by and large unaffected (either the same or a tad faster within noise range).
Here are the results running on OpenSSL 1.0.2d on osx (12 cores, 32G RAM)
=== base
infer binary: 26193088 bytes
compile time: 40s
capture:
```lang=text
real 1m7.513s
user 3m11.437s
sys 0m55.236s
```
analysis:
```lang=text
real 5m41.580s
user 61m37.855s
sys 1m12.870s
```
Memory profile:
```lang=json
{
...
"minor_gb": 0.1534719169139862,
"promoted_gb": 0.0038930922746658325,
"major_gb": 0.4546157643198967,
"allocated_gb": 0.6041945889592171,
"minor_collections": 78,
"major_collections": 23,
"compactions": 7,
"top_heap_gb": 0.07388687133789062,
"stack_kb": 0.3984375,
"minor_heap_kb": 8192.0,
...
}
```
=== flambda with stock options (no `-Oclassic`, just the same flags as base)
Exactly the same as base.
=== flambda `-O3`
infer binary: 56870376 bytes (2.17x bigger)
compile time: 191s (4.78x slower)
capture is the same as base:
```lang=text
real 1m9.203s
user 3m12.242s
sys 0m58.905s
```
analysis is ~20% wallclock time faster, ~25% CPU time faster:
```lang=text
real 4m32.656s
user 46m43.987s
sys 1m2.424s
```
memory usage is a bit lower too:
```lang=json
{
...
"minor_gb": 0.11583046615123749, // 75% of previous
"promoted_gb": 0.00363825261592865, // 93% of previous
"major_gb": 0.45415670424699783, // about same
"allocated_gb": 0.5663489177823067, // 94% of previous
"minor_collections": 73,
"major_collections": 22,
"compactions": 7,
"top_heap_gb": 0.07165145874023438,
"stack_kb": 0.3359375,
"minor_heap_kb": 8192.0,
...
}
```
=== flambda `-O2`
Not nearly as exciting as `-O3`, but the compilation cost is still quite high:
infer: 37826856 bytes
compilation of infer: 100s
Capture and analysis timings are mostly the same as base.
Reviewed By: jberdine
Differential Revision: D4867979
fbshipit-source-id: 99230b7
Summary:
This can be a long-running step and it's useful to know how long it took. We
already dump some statistics on stderr after merging is done, this just adds
one more line.
Reviewed By: mbouaziz
Differential Revision: D5833580
fbshipit-source-id: 70e19ab
Summary:
Simple instance of the problem: analyzing the following program times out.
```
#include <tuple>
void foo() {
std::tuple<std::tuple<int>> x;
}
```
Replacing `std::tuple<std::tuple<int>>` by `std::tuple<int>` makes the analysis
terminate.
In the AST, both tuple<tuple<int>> and tuple<int> have the same template
specialization type: "Pack" (which means we're supposed to go look into the
arguments of the template to get their values). This is not information enough
and that's the plugin fault.
On the backend side, this means that two types have the same Typ.Name.t, namely
"std::tuple<_>", so they collide in the tenv. The definition of
tuple<tuple<int>> is the one making it into the tenv. One of the fields of the
corresponding CxxRecord is of type "tuple<int>", which we see as the same
"tuple<_>", which causes the loop.
update-submodule: facebook-clang-plugins
Reviewed By: mbouaziz
Differential Revision: D5775840
fbshipit-source-id: 0528604
Summary:
Sort the complete set of warnings by everything except procname, then de-duplicate.
This scheme prevents reporting identical error messages on the same line/same file.
This is important for avoiding duplicate reports on multiple instantiations of the same template.
Reviewed By: jberdine
Differential Revision: D5819467
fbshipit-source-id: 984f47f
Summary: The resolution was previously only happening for constructors, but calls to private methods or to `super` are also neither static calls nor virtual calls. In this case, the resolution logic should be the same as for constructors.
Reviewed By: sblackshear
Differential Revision: D5830376
fbshipit-source-id: 9b56f80
Summary:
The reporting phases iterates over each procedure summary and print all the issues from each procedure.
That's nice because we don't have to build a big list of the issues in-memory, but it's not so nice if you want to ouput the reports in a certain order or de-duplicate them.
This diff builds the in-memory list and outputs the issues afterward. By itself, this isn't very useful. But in the near future it will allow us to:
- Group all of the issues from the same file (finally!!!)
- Get rid of duplicate issues on multiple instantiations of the same C++ template
- Probably other cool stuff too
Reviewed By: jeremydubreil, mbouaziz
Differential Revision: D5816646
fbshipit-source-id: 799bcd0
Summary:
The clang frontend uses `assert false` for unimplemented features that should
abort method translations, as well as for genuine internal errors.
Distinguishing between the two, we can fail hard on the latter and not the
former.
1. This introduces a new exception `Unimplemented` that is used instead of `assert false` where appropriate.
2. Changed some other cases into `die ...` (when there's an error message to display)
3. Wherever a path in the code that we assumed to be unreachable was observed reachable, we now raise `IncorrectAssumption`. These should be fixed, but the fixes are not obvious.
Reviewed By: sblackshear
Differential Revision: D5784384
fbshipit-source-id: 61b55af
Summary:
In the most recent version of clang plugin, lambda captures were moved from `LambdaExpr` to `CXXRecordDecl`. Updating infer to reflect this change.
update-submodule: facebook-clang-plugins
Reviewed By: jvillard
Differential Revision: D5823034
fbshipit-source-id: dd5fc45
Summary: This will run infer with `--no-keep-going` by default in our tests.
Reviewed By: sblackshear
Differential Revision: D5814237
fbshipit-source-id: c1e1a4e
Summary:
"Running as clang" was its own infer subcommand. That's not terribly good
because it makes it hard to specify another subcommand, and in particular it
broke the `compile` subcommand in some integrations because "running as clang"
would always do capture. The cmake tests required to run with `--keep-going`
because of this.
Instead of having its own fake subcommand, simply add a new boolean in the
config for "infer runs as clang", as we do for javac already (used in the mvn
integration).
Also make logging of the environment better.
Reviewed By: jberdine
Differential Revision: D5813986
fbshipit-source-id: 72b96cd
Summary: This is to prevent test failures to happen whenchanging the code a little.
Reviewed By: sblackshear
Differential Revision: D5815349
fbshipit-source-id: 8516102
Summary: Destroying local variables that are out of scope after `continue`.
Reviewed By: jberdine
Differential Revision: D5804120
fbshipit-source-id: 638cff5
Summary: I often find myself needing a generic `Pp.seq` where I can specify the separator.
Reviewed By: jeremydubreil
Differential Revision: D5803915
fbshipit-source-id: fb8d30d
Summary: That match branch is Java-only but could be reached with a C++ method, causing a crash.
Reviewed By: jberdine, mbouaziz
Differential Revision: D5814041
fbshipit-source-id: 6b1b501
Summary: It's useful to be able to configure both max depth and max width
Reviewed By: jeremydubreil
Differential Revision: D5801567
fbshipit-source-id: 0138cd7
Summary:
Races on the internal implementation of data structures defined in
system headers are currently not detected, since the memory accesses
are in procedures that are not analyzed.
This diff adds models for a few std::map operations that indicate if
they read or write the underlying representation.
Reviewed By: da319
Differential Revision: D5804293
fbshipit-source-id: 55ff28c
Summary: Try to preserve the original backtrace. Introduce `reraise` in the global namespace.
Reviewed By: jberdine
Differential Revision: D5804121
fbshipit-source-id: 0947a47
Summary:
This makes it much easier to read infer logs coming from these subprocesses
that use a sub-results dir.
Reviewed By: jberdine
Differential Revision: D5777783
fbshipit-source-id: f074536
Summary: Destroying local variables that are out of scope after `break`.
Reviewed By: jberdine
Differential Revision: D5764647
fbshipit-source-id: a7e06ae
Summary: The successor node of `continue` was not correct inside the `do while`.
Reviewed By: sblackshear
Differential Revision: D5769578
fbshipit-source-id: d7b0843
Summary:
Read the documentation and it doesn't seem like these functions are guaranteed to choose the same value in different runs.
I hypothesize that these may be the source of flakiness in the thread-safety tests/smoke tests.
Reviewed By: jvillard
Differential Revision: D5794384
fbshipit-source-id: 02b7a96
Summary: With this diff, the analysis trace will jump to the definition of the skipped methods when the location is known. This is especially useful when the analysis is relying on the method annotations.
Reviewed By: sblackshear
Differential Revision: D5783428
fbshipit-source-id: 561b739
Summary: The code was previously inconsistent as passing the option was using an empty list of instructions to translate the code but still using the original JBIR representation to connect the nodes.
Reviewed By: sblackshear
Differential Revision: D5795903
fbshipit-source-id: 88c0e64
Summary:
The behaviour of infer was observed to be different in 4.04.2 vs (4.05.0 or
4.04.2+flambda). Investigating further, the behaviour is also different in byte
vs native versions of infer under 4.04.2. Looking at the Changelog of 4.05.0
(thanks mbouaziz), there are fixes for inconsistent behaviours between byte
and native relative to evaluation order.
Lots of debugging later, this 1 line patch is born. Also use `List.concat_map`
instead of re-implementing it.
Reviewed By: jberdine
Differential Revision: D5793809
fbshipit-source-id: 374fb4c
Summary: With Logging.exit you have more control of the code that invokes exit, for example when forking and running certain functions that may in turn invoke exit, and you want to handle the execution flow differently - like invoking certain callbacks before exiting, or not exiting at all.
Reviewed By: jvillard
Differential Revision: D5746914
fbshipit-source-id: 596fba1
Summary: This will prevent trivial errors like misspelling node names, e.g. when those are typed as part of the operators `IN-NODE <node-name>` or `HOLDS-IN-NODE <node-name>`
Reviewed By: dulmarod
Differential Revision: D5680411
fbshipit-source-id: e241c1c
Summary: This should be equivalent as far as bug finding is concerned, but will allow the analysis traces to jump into the skipped methods.
Reviewed By: sblackshear
Differential Revision: D5778245
fbshipit-source-id: 3fd061f
Summary:
We supported globals as sources before, but we did so by allowing ClangTrace etc. to match against any access path in the footprint of the trace.
This is very powerful/flexible, but it's ultimately not a good idea because it leads to traces that are hard to read.
This is because a footprint source doesn't have any information about its provenance: we might know that the value came from a global, but we don't know where the read occurred.
The mechanism for handling procedure calls as sources already knows how to solve this problem.
This diff implements globals as sources as a special case of procedure call sources instead.
This will give us much nicer traces with full provenance of the read from the global.
Reviewed By: mbouaziz
Differential Revision: D5772299
fbshipit-source-id: 491ae81
Summary: This is almost equivalent to the previous one except in the case where new and old are both undefined: before, we would just pick "old", but now we pick the biggest according to their source files. I think the previous behaviour was a bug because it was more non-deterministic.
Reviewed By: jberdine
Differential Revision: D5649481
fbshipit-source-id: aeb527d
Summary: It is not clear to me what the removed code was for in the first place. Basically, it was replacing the pure part of propositions in a semantically equivalent way, e.g. replacing `a = b /\ Attribute(a)` to `a = b /\ Attribute(b)`, and `a = b /\ Attribute(b)` to `a = b /\ Attribute(a)`.
Reviewed By: jberdine
Differential Revision: D5657366
fbshipit-source-id: 93cd9e0
Summary: The list of fields of a Java object in SIL is the list of fields declared in the class plus all the fields declared in the the super classes. It turns out that we were missing the fields declared in the implemented interfaces.
Reviewed By: sblackshear
Differential Revision: D5720386
fbshipit-source-id: d65c9de
Summary:
Calling `Exn.backtrace` doesn't give us the current backtrace, only the one of
the latest exception raised. Change the logic of toplevel exception catching to
print the right backtrace.
Also, do not use `Format` to print from `Config` as `Logging` may have messed
with it already.
Finally, throw in some colours!
Reviewed By: jberdine
Differential Revision: D5764825
fbshipit-source-id: cd51688
Summary: Add procedure where error is triggered (if available) + bad data to messages.
Reviewed By: jvillard
Differential Revision: D5756398
fbshipit-source-id: a16f7cf
Summary:
When a lambda has an `auto` parameter, the inferred type of the parameter because part of the name.
Our heuristic for identifying lambda was checking if the lambda's name was exactly `operator()`, which won't catch this case.
Reviewed By: jvillard
Differential Revision: D5753323
fbshipit-source-id: 85ff75a
Summary:
Not translating these properly was causing false positives for the dead store analysis in cases like
```
int i = 0;
return [j = i]() { return j; }();
```
Reviewed By: da319
Differential Revision: D5731562
fbshipit-source-id: ae79ac8
Summary:
Keep track of generated files better and make sure we nuke them on `make clean`.
Also get rid of a few unused variables in infer/src/Makefile.
Reviewed By: sblackshear
Differential Revision: D5754553
fbshipit-source-id: b5fca41
Summary: We inject destructor calls of base classes inside destructor bodies after the destructor calls of fields.
Reviewed By: dulmarod
Differential Revision: D5745499
fbshipit-source-id: 90745ec
Summary: Refactoring destructor translation to take type pointer instead of qual type as a parameter, as we only have type pointers for base classes.
Reviewed By: dulmarod
Differential Revision: D5745490
fbshipit-source-id: 121f492
Summary: We used to crash whenever we hit these. The simple translation implemented here is not particularly inspiring, but it is better than crashing.
Reviewed By: jvillard
Differential Revision: D5702095
fbshipit-source-id: 3795d43
Summary: This appears to be interchangeable with `Config.debug`, so let's kill it.
Reviewed By: jvillard
Differential Revision: D5742685
fbshipit-source-id: d390b6b
Sam Blackshear
Jeremy Dubreil
Jules Villard
Varun Arora
Dulma Churchill
Mehdi Bouaziz
Daiva Naudziuniene
Tony Hu
Martino Luca
Radu Grigore
Vladimir Silyaev
Josh Berdine
Dino Distefano
Peter O'Hearn
Peter Goldsborough
Sungkeun Cho
Vivek Dhiman
Philip Cohn-Cort
Angel (scout387)
Stephan Walters
anuolu
Dominic Cooney
Anders Papitto
Kyriakos Nikolaos Gkorogiannis